Stroke is the second leading cause of death and the third leading cause of long-term morbidity worldwide. Among all causes of stroke, intracranial atherosclerotic disease (ICAD) represents 9-15% of strokes in the US, and the percentage is higher in patients of Asian and other non-Caucasian ethnicity. The detection of ICAD currently relies on stenosis measurement and likely underestimates the burden of this disease. Vessel wall imaging of the extracranial (EC) carotid artery is well established. The core technology for characterizing key atherosclerotic plaque features requires the use of multiple contrast weighted (pre, post contrast T1, T2 weighted and luminal) magnetic resonance imaging (MRI) with effective signal suppression of flowing blood. While there is growing expertise with vessel wall imaging in EC carotid artery atherosclerosis, its role in ICAD remains to be studied. The technical challenges for intracranial vessel wall (IVW) imaging are the small caliber and tortuous nature of the intracranial arteries, the surrounding cerebrospinal fluid (CSF) that has similar contrast as the vessel wall itself, the close proximity of the arteries to the skull base and paranasal sinuses that can obscure the vessels, and the large area that needs to be covered to identify multiple lesion sites. In this proposal, we plan to 1) develop a multi-contrast 3D high resolution (HR) IVW MRI technique that consists of time efficient imaging sequences with a novel method for outer wall boundary detection; 2) develop image processing tools to characterize lesion size, distribution, main composition features and luminal surface condition-collectively named 3D wall imaging (3D-WALLI); 3) establish a scoring system with a stronger association with clinical symptoms and ischemic brain lesions than stenosis; and 4) study the score's power to predict secondary stroke and new ischemic lesions. The successful achievement of these aims will substantially expand our understanding of the relationship between ICAD vessel wall characteristics and risk of stroke, and 3D-WALLI will help define subgroups at higher risk which may represent a target population for future trials of novel interventions. Given the large global burden of this ischemic stroke subtype, the findings will have broad applicability.